This is Kara’s second attempt at looking at the Taves data by site, looking only at participant’s endorsements of whether or not they had had certain experiences (not their assessment of what they meant, etc.).
Below are plots of exploratory factor analyses (one per site) using polychoric correlations (since we’re working with yes/no responses), an orthogonal rotation (“varimax,” which forces factors to be uncorrelated with each other - this is what Ann has used in previous work), and a “maximum likelihoood” factoring method (for comparison with Ann’s previous work).
There are many different ways to decide how many factors to retain, and different methods make different suggestions. For now, I’ve assessed 4 ways of deciding this:
To help us think through this, I’ve included a table of the items that are more strongly positively related to each factor, for each site, for each of these solutions. (This is redundant with the plot for that solution, but easier to read.)
Below the factor analyses are plots of item-level correlations, organized by hierarchical clustering. What you’re looking for are patches of red - anything inside the patch could be thought of as a “cluster.”
This is all as much art as science, but it’s a place to start!
Notes: per our conversation with Nikki, we are dropping one question (#53), which was a repeated question in all sites except for China.
Joining, by = "taves_subj"
Minimizing BIC suggests retaining 8 factors:
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.84 |
|
0.67 |
|
0.60 |
|
0.59 |
| Factor 2 | |
|
0.85 |
|
0.35 |
|
0.34 |
|
0.32 |
| Factor 3 | |
|
0.71 |
|
0.65 |
|
0.52 |
|
0.50 |
| Factor 4 | |
|
0.62 |
|
0.60 |
|
0.59 |
|
0.59 |
| Factor 5 | |
|
0.75 |
|
0.74 |
|
0.63 |
|
0.59 |
| Factor 6 | |
|
0.92 |
|
0.66 |
|
0.45 |
|
0.39 |
| Factor 7 | |
|
0.57 |
|
0.57 |
|
0.56 |
|
0.51 |
| Factor 8 | |
|
0.66 |
|
0.62 |
|
0.55 |
|
0.49 |
Kara’s home-brewed approach suggests retaining 3 factors:
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.82 |
|
0.69 |
|
0.65 |
|
0.63 |
|
0.59 |
|
0.55 |
|
0.54 |
|
0.52 |
|
0.51 |
|
0.51 |
| Factor 2 | |
|
0.82 |
|
0.62 |
|
0.58 |
|
0.57 |
|
0.55 |
|
0.53 |
|
0.52 |
|
0.52 |
|
0.51 |
|
0.51 |
| Factor 3 | |
|
0.61 |
|
0.61 |
|
0.61 |
|
0.57 |
|
0.54 |
|
0.52 |
|
0.52 |
|
0.52 |
|
0.49 |
|
0.49 |
42 cells were adjusted for 0 values using the correction for continuity. Examine your data carefully.Matrix was not positive definite, smoothing was done
Kara’s take on Ann’s home-brewed approach suggests retaining 9 factors:
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.75 |
|
0.65 |
|
0.46 |
| Factor 2 | |
|
0.91 |
|
0.58 |
|
0.56 |
| Factor 3 | |
|
0.85 |
|
0.56 |
|
0.49 |
| Factor 4 | |
|
0.63 |
|
0.57 |
|
0.52 |
| Factor 5 | |
|
0.60 |
|
0.56 |
|
0.49 |
| Factor 6 | |
|
0.80 |
|
0.65 |
|
0.59 |
| Factor 7 | |
|
0.77 |
|
0.53 |
|
0.52 |
| Factor 8 | |
|
0.77 |
|
0.62 |
|
0.58 |
| Factor 9 | |
|
0.70 |
|
0.63 |
|
0.62 |
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.75 |
|
0.74 |
|
0.63 |
|
0.59 |
| Factor 2 | |
|
0.62 |
|
0.60 |
|
0.59 |
|
0.59 |
| Factor 3 | |
|
0.84 |
|
0.67 |
|
0.60 |
|
0.59 |
| Factor 4 | |
|
0.57 |
|
0.57 |
|
0.56 |
|
0.51 |
| Factor 5 | |
|
0.71 |
|
0.65 |
|
0.52 |
|
0.50 |
| Factor 6 | |
|
0.66 |
|
0.62 |
|
0.55 |
|
0.49 |
| Factor 7 | |
|
0.92 |
|
0.66 |
|
0.45 |
|
0.39 |
| Factor 8 | |
|
0.85 |
|
0.35 |
|
0.34 |
|
0.32 |
Correlation method: 'pearson'
Missing treated using: 'pairwise.complete.obs'
Minimizing BIC suggests retaining 8 factors:
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.87 |
|
0.65 |
|
0.37 |
|
0.37 |
| Factor 2 | |
|
0.87 |
|
0.35 |
|
0.33 |
|
0.29 |
| Factor 3 | |
|
0.80 |
|
0.68 |
|
0.53 |
|
0.52 |
| Factor 4 | |
|
0.63 |
|
0.61 |
|
0.53 |
|
0.50 |
| Factor 5 | |
|
0.72 |
|
0.67 |
|
0.60 |
|
0.56 |
| Factor 6 | |
|
0.66 |
|
0.62 |
|
0.61 |
|
0.58 |
| Factor 7 | |
|
0.57 |
|
0.54 |
|
0.52 |
|
0.49 |
| Factor 8 | |
|
0.68 |
|
0.61 |
|
0.55 |
|
0.53 |
Kara’s home-brewed approach suggests retaining 2 factors:
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.74 |
|
0.70 |
|
0.67 |
|
0.66 |
|
0.64 |
|
0.64 |
|
0.64 |
|
0.64 |
|
0.63 |
|
0.60 |
|
0.59 |
|
0.58 |
|
0.56 |
|
0.55 |
|
0.54 |
|
0.53 |
| Factor 2 | |
|
0.70 |
|
0.67 |
|
0.63 |
|
0.63 |
|
0.63 |
|
0.61 |
|
0.59 |
|
0.57 |
|
0.57 |
|
0.56 |
|
0.54 |
|
0.52 |
|
0.52 |
|
0.51 |
|
0.51 |
|
0.51 |
Matrix was not positive definite, smoothing was done
Kara’s take on Ann’s home-brewed approach suggests retaining 6 factors:
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.88 |
|
0.62 |
|
0.45 |
|
0.42 |
|
0.36 |
| Factor 2 | |
|
0.82 |
|
0.67 |
|
0.57 |
|
0.56 |
|
0.51 |
| Factor 3 | |
|
0.70 |
|
0.65 |
|
0.53 |
|
0.49 |
|
0.45 |
| Factor 4 | |
|
0.90 |
|
0.63 |
|
0.60 |
|
0.52 |
|
0.52 |
| Factor 5 | |
|
0.66 |
|
0.62 |
|
0.59 |
|
0.59 |
|
0.56 |
| Factor 6 | |
|
0.54 |
|
0.50 |
|
0.50 |
|
0.47 |
|
0.46 |
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.66 |
|
0.62 |
|
0.61 |
|
0.58 |
| Factor 2 | |
|
0.80 |
|
0.68 |
|
0.53 |
|
0.52 |
| Factor 3 | |
|
0.72 |
|
0.67 |
|
0.60 |
|
0.56 |
| Factor 4 | |
|
0.63 |
|
0.61 |
|
0.53 |
|
0.50 |
| Factor 5 | |
|
0.68 |
|
0.61 |
|
0.55 |
|
0.53 |
| Factor 6 | |
|
0.57 |
|
0.54 |
|
0.52 |
|
0.49 |
| Factor 7 | |
|
0.87 |
|
0.65 |
|
0.37 |
|
0.37 |
| Factor 8 | |
|
0.87 |
|
0.35 |
|
0.33 |
|
0.29 |
Correlation method: 'pearson'
Missing treated using: 'pairwise.complete.obs'
Minimizing BIC suggests retaining 8 factors:
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.71 |
|
0.68 |
|
0.64 |
|
0.63 |
| Factor 2 | |
|
0.86 |
|
0.31 |
|
0.31 |
|
0.26 |
| Factor 3 | |
|
0.94 |
|
0.53 |
|
0.48 |
|
0.41 |
| Factor 4 | |
|
0.90 |
|
0.42 |
|
0.41 |
|
0.36 |
| Factor 5 | |
|
0.83 |
|
0.74 |
|
0.74 |
|
0.73 |
| Factor 6 | |
|
0.68 |
|
0.54 |
|
0.51 |
|
0.47 |
| Factor 7 | |
|
0.68 |
|
0.38 |
|
0.36 |
|
0.29 |
| Factor 8 | |
|
0.82 |
|
0.81 |
|
0.71 |
|
0.69 |
Kara’s home-brewed approach suggests retaining 3 factors:
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.82 |
|
0.74 |
|
0.72 |
|
0.70 |
|
0.66 |
|
0.66 |
|
0.65 |
|
0.64 |
|
0.63 |
|
0.59 |
| Factor 2 | |
|
0.77 |
|
0.75 |
|
0.73 |
|
0.72 |
|
0.69 |
|
0.67 |
|
0.67 |
|
0.66 |
|
0.64 |
|
0.63 |
| Factor 3 | |
|
0.79 |
|
0.68 |
|
0.63 |
|
0.57 |
|
0.57 |
|
0.54 |
|
0.52 |
|
0.52 |
|
0.51 |
|
0.49 |
39 cells were adjusted for 0 values using the correction for continuity. Examine your data carefully.Matrix was not positive definite, smoothing was done
Kara’s take on Ann’s home-brewed approach suggests retaining 9 factors:
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.82 |
|
0.80 |
|
0.72 |
| Factor 2 | |
|
0.86 |
|
0.37 |
|
0.31 |
| Factor 3 | |
|
0.94 |
|
0.53 |
|
0.47 |
| Factor 4 | |
|
0.82 |
|
0.76 |
|
0.74 |
| Factor 5 | |
|
0.59 |
|
0.35 |
|
0.35 |
| Factor 6 | |
|
0.74 |
|
0.54 |
|
0.51 |
| Factor 7 | |
|
0.55 |
|
0.37 |
|
0.33 |
| Factor 8 | |
|
0.76 |
|
0.67 |
|
0.62 |
| Factor 9 | |
|
0.79 |
|
0.40 |
|
0.37 |
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.83 |
|
0.74 |
|
0.74 |
|
0.73 |
| Factor 2 | |
|
0.82 |
|
0.81 |
|
0.71 |
|
0.69 |
| Factor 3 | |
|
0.71 |
|
0.68 |
|
0.64 |
|
0.63 |
| Factor 4 | |
|
0.94 |
|
0.53 |
|
0.48 |
|
0.41 |
| Factor 5 | |
|
0.68 |
|
0.54 |
|
0.51 |
|
0.47 |
| Factor 6 | |
|
0.90 |
|
0.42 |
|
0.41 |
|
0.36 |
| Factor 7 | |
|
0.68 |
|
0.38 |
|
0.36 |
|
0.29 |
| Factor 8 | |
|
0.86 |
|
0.31 |
|
0.31 |
|
0.26 |
Correlation method: 'pearson'
Missing treated using: 'pairwise.complete.obs'
Minimizing BIC suggests retaining 8 factors:
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.66 |
|
0.62 |
|
0.58 |
|
0.56 |
| Factor 2 | |
|
0.51 |
|
0.49 |
|
0.42 |
|
0.41 |
| Factor 3 | |
|
0.69 |
|
0.67 |
|
0.54 |
|
0.52 |
| Factor 4 | |
|
0.71 |
|
0.58 |
|
0.55 |
|
0.54 |
| Factor 5 | |
|
0.67 |
|
0.54 |
|
0.54 |
|
0.53 |
| Factor 6 | |
|
0.64 |
|
0.59 |
|
0.58 |
|
0.52 |
| Factor 7 | |
|
0.65 |
|
0.57 |
|
0.55 |
|
0.49 |
| Factor 8 | |
|
0.77 |
|
0.66 |
|
0.56 |
|
0.50 |
Kara’s home-brewed approach suggests retaining 2 factors:
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.84 |
|
0.67 |
|
0.61 |
|
0.60 |
|
0.59 |
|
0.56 |
|
0.49 |
|
0.48 |
|
0.48 |
|
0.48 |
|
0.47 |
|
0.46 |
|
0.46 |
|
0.45 |
|
0.45 |
|
0.44 |
| Factor 2 | |
|
0.67 |
|
0.66 |
|
0.64 |
|
0.60 |
|
0.60 |
|
0.55 |
|
0.55 |
|
0.55 |
|
0.54 |
|
0.54 |
|
0.51 |
|
0.51 |
|
0.49 |
|
0.49 |
|
0.48 |
|
0.46 |
57 cells were adjusted for 0 values using the correction for continuity. Examine your data carefully.Matrix was not positive definite, smoothing was done
Kara’s take on Ann’s home-brewed approach suggests retaining 10 factors:
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.68 |
|
0.66 |
|
0.66 |
| Factor 2 | |
|
0.65 |
|
0.60 |
|
0.43 |
| Factor 3 | |
|
0.88 |
|
0.44 |
|
0.41 |
| Factor 4 | |
|
0.70 |
|
0.47 |
|
0.27 |
| Factor 5 | |
|
0.88 |
|
0.53 |
|
0.48 |
| Factor 6 | |
|
0.74 |
|
0.60 |
|
0.53 |
| Factor 7 | |
|
0.75 |
|
0.59 |
|
0.53 |
| Factor 8 | |
|
0.65 |
|
0.63 |
|
0.55 |
| Factor 9 | |
|
0.78 |
|
0.68 |
|
0.48 |
| Factor 10 | |
|
0.76 |
|
0.62 |
|
0.56 |
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.77 |
|
0.66 |
|
0.56 |
|
0.50 |
| Factor 2 | |
|
0.66 |
|
0.62 |
|
0.58 |
|
0.56 |
| Factor 3 | |
|
0.69 |
|
0.67 |
|
0.54 |
|
0.52 |
| Factor 4 | |
|
0.71 |
|
0.58 |
|
0.55 |
|
0.54 |
| Factor 5 | |
|
0.67 |
|
0.54 |
|
0.54 |
|
0.53 |
| Factor 6 | |
|
0.65 |
|
0.57 |
|
0.55 |
|
0.49 |
| Factor 7 | |
|
0.64 |
|
0.59 |
|
0.58 |
|
0.52 |
| Factor 8 | |
|
0.51 |
|
0.49 |
|
0.42 |
|
0.41 |
Correlation method: 'pearson'
Missing treated using: 'pairwise.complete.obs'
Minimizing BIC suggests retaining 8 factors:
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.95 |
|
0.44 |
|
0.40 |
|
0.33 |
| Factor 2 | |
|
0.77 |
|
0.67 |
|
0.58 |
|
0.57 |
| Factor 3 | |
|
0.92 |
|
0.45 |
|
0.41 |
|
0.37 |
| Factor 4 | |
|
0.66 |
|
0.60 |
|
0.59 |
|
0.58 |
| Factor 5 | |
|
0.77 |
|
0.70 |
|
0.59 |
|
0.54 |
| Factor 6 | |
|
0.72 |
|
0.62 |
|
0.58 |
|
0.44 |
| Factor 7 | |
|
0.61 |
|
0.60 |
|
0.54 |
|
0.48 |
| Factor 8 | |
|
0.79 |
|
0.57 |
|
0.47 |
|
0.46 |
Kara’s home-brewed approach suggests retaining 2 factors:
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.68 |
|
0.64 |
|
0.63 |
|
0.63 |
|
0.63 |
|
0.62 |
|
0.61 |
|
0.61 |
|
0.60 |
|
0.58 |
|
0.58 |
|
0.55 |
|
0.55 |
|
0.52 |
|
0.47 |
|
0.46 |
| Factor 2 | |
|
0.66 |
|
0.63 |
|
0.62 |
|
0.59 |
|
0.57 |
|
0.56 |
|
0.56 |
|
0.54 |
|
0.53 |
|
0.51 |
|
0.49 |
|
0.48 |
|
0.46 |
|
0.45 |
|
0.45 |
|
0.45 |
Matrix was not positive definite, smoothing was done
Kara’s take on Ann’s home-brewed approach suggests retaining 12 factors:
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.95 |
|
0.58 |
| Factor 2 | |
|
0.64 |
|
0.60 |
| Factor 3 | |
|
0.62 |
|
0.61 |
| Factor 4 | |
|
0.68 |
|
0.55 |
| Factor 5 | |
|
0.92 |
|
0.41 |
| Factor 6 | |
|
0.86 |
|
0.62 |
| Factor 7 | |
|
0.91 |
|
0.45 |
| Factor 8 | |
|
0.84 |
|
0.52 |
| Factor 9 | |
|
0.94 |
|
0.50 |
| Factor 10 | |
|
0.72 |
|
0.71 |
| Factor 11 | |
|
0.77 |
|
0.51 |
| Factor 12 | |
|
0.68 |
|
0.62 |
Joining, by = "capacity"
Joining, by = "factor"
| capacity | loading |
|---|---|
| Factor 1 | |
|
0.77 |
|
0.67 |
|
0.58 |
|
0.57 |
| Factor 2 | |
|
0.77 |
|
0.70 |
|
0.59 |
|
0.54 |
| Factor 3 | |
|
0.66 |
|
0.60 |
|
0.59 |
|
0.58 |
| Factor 4 | |
|
0.61 |
|
0.60 |
|
0.54 |
|
0.48 |
| Factor 5 | |
|
0.95 |
|
0.44 |
|
0.40 |
|
0.33 |
| Factor 6 | |
|
0.79 |
|
0.57 |
|
0.47 |
|
0.46 |
| Factor 7 | |
|
0.72 |
|
0.62 |
|
0.58 |
|
0.44 |
| Factor 8 | |
|
0.92 |
|
0.45 |
|
0.41 |
|
0.37 |
Correlation method: 'pearson'
Missing treated using: 'pairwise.complete.obs'